Summary
The manuscript has been revised based on comments from two reviewers including myself. I am satisfied with most of the revision. I think the results are very interesting and add insight to the role of latent heating in the midlatitude circulation.
Besides a few minor comments, I still have two major suggestions. One suggestion is for an additional figure that could aid the interpretation of the results (see major comment 1). The second suggestion (major comment 2) is regarding the physical interpretation of the difference the authors find between the effect of latent heating on baroclinicity for the averaged and instantaneous flows. I don’t disagree with the interpretation given in the manuscript, but I think that using other well-known concepts from the literature could add to the contribution of the paper to the theoretical understanding of the role of latent heating in the midlatitude circulation.
Major comments
1) As explained in lines 124-127, the results presented in this paper are calculated by binning the data and averaging over all members in each bin. It could be that the sample size varies greatly between bins. I think that adding a figure showing the sample size for each bin (for the 3h-Z, 1d-Z, 10d-Z, 3h-2D, 10d-2D and for the 3h-2D sorted by jet direction) would help to interpret the results better. It is important to see which cases are common and which are rare. For example, in lines 140-141 the authors mention that the patterns found at high latitudes in the 10d-Z analysis might not be robust due to the small sample size.
2) As implied in the title and in many places in the manuscript (e.g. lines 209-212, 256-262 and 272-278), the authors argue that latent heating reinforces baroclinicity and that averaging over longitude and time obscures this picture. I agree that this statement is consistent with the results, but I think it gives a misleading impression for the role of latent heating. Looking for example at the lower panel of figure 1b, it is clear that latent heating occurs inside the warm conveyor belt of a frontal cyclone. The “jet” in this case is the upper level northward wind inside the cyclone, which is maximal slightly westward of the latent heating that occurs in the mid-troposphere. It is true that strictly speaking this reinforces the baroclinicity inside the cyclone. However, the term “baroclinicity” usually refers to the background (averaged) flow, which may become baroclinically unstable and allow for further growth of eddies (mostly cyclones). Here I think this local baroclinicity represents the available potential energy of the cyclone and not the background flow. One can think of it in the context of the Lorenz energy cycle: mean available potential energy (MAPE) is converted to eddy available potential energy (EAPE) which is then converted to eddy kinetic energy (EKE). I would argue that the results of this paper demonstrate that latent heating reduces MAPE and increases EAPE. Therefore it plays a role in the energy transfer from the mean flow to the eddies. This interpretation is based on the following results:
a. Latent heating is located eastward of the upper level northward instantaneous jet (Figures 5g, 7c, 9, and the example in Figure 1).
b. The regions of strong latent heating are characterized by upward wind, large-scale precipitation, positive temperature anomaly and high specific humidity (Figures 5-7).
c. Latent heating is concentrated on the poleward side of the zonally- and temporally-averaged jet (Figure 3j,k,l).
Results a and b fit well with the characteristics of frontal cyclones and warm conveyor belts. Inside the cyclone there is a positive correlation between temperature and upward wind (w’T’>0) and temperature and northward wind (v’T>0), as well as specific humidity and upward or northward wind (w’q’>0, v’q’>0). The positive correlation between temperature and northward wind (v’T’>0) is necessary for the conversion of MAPE to EAPE and the positive correlation between temperature upward wind (w’T’>0) is necessary for the conversion of EAPE to EKE (see Lorenz, 1955; and a recent example in Okajima et al. 2022). Latent heating then contributes to this temperature anomaly and to the conversion of MAPE to EAPE. Note that in this paper the “jet” is measured at a higher level than the latent heating so that if the flow inside the cyclone is tilted westward with height (as in a characteristic baroclinically unstable eddy), then the upper level northward wind would be slightly westward of the mid-tropospheric northward wind and the positive temperature (and latent heating) anomaly that is positively correlated with it. Also note that the opposite patterns of upward/downward motion between the instantaneous and averaged jets (lines 182-184) are consistent with the differences between the Ferrel circulation and the Lagrangian-mean circulation (as approximated by the Transformed Eulerian Mean).
The fact that latent heating is concentrated on the poleward side of the zonally averaged jet (figure 3) demonstrates that it acts to decrease MAPE. Therefore, I don’t think that saying that latent heating reinforces the baroclinicity delivers the right impression for those thinking in terms of wave-mean flow interaction.
My suggestion is to use the well-known concepts of the Lorenz energy cycle to explain the role of latent heating in the midlatitude circulation. The authors can consider whether to accept this suggestion or not.
Minor comments
1) Line 48: two comments: (a) “explore the extent to which” instead of “explore to the extent which”. (b) “… occurs on the cold or warm flank of each jet definition” instead of “… occurs on the poleward or equatorward flank of each jet definition”. For the instantaneous jet its really mostly the westward vs eastward flank.
2) Line 74: “only allow for one jet latitude for each time step” (add “for each”).
3) Line 79: since you removed the hat above n in equation (1), it needs to be removed here as well.
4) The paragraph in lines 83-91: The parameter K is not defined. Shouldn’t it have units? It is written here as if it is a dimensionless parameter, but it should have units of sec^(-2) if n has units of length. Also there is some confusion with the sign. If you use it to detect a maximum point then K should be negative and the inequality in line 86 should be opposite (the second derivative is lower than some negative threshold). In line 96 it has a positive value, which seems inconsistent.
5) Line 91: “de tails” – change to “details” (delete the space).
6) Lines 99-100: “Fig. 1a shows…” (add “a”) and remove “(Fig. 1a)” at the end of the sentence.
7) Lines 118-119: Here it should be “left side” and “right side” instead of “cold flanks” and “warm flanks”. It can be explained that the left side is cold and the right side is warm, due to thermal wind balance, and then the terms “cold flanks” and “warm flanks” can be used for the rest of the paper, but since this sentence describes the geometry of the analysis I think the terms “right” and “left” should be used here.
8) Lines 185-186 and 188-189: “ascent appears to be mainly a function of latitude” and “the specific humidity is… mostly a function of latitude” – it should be “mostly a function of the jet latitude”, because if it were a function of latitude, it would have had a diagonal structure, as in figure 3.
9) Caption of figure 5: I think the last sentence should be moved to the methods section.
10) Line 223: “Figure 7 only shows” instead of “only show”.
11) Line 237: “Appendix A” instead of “Sect. A”.
12) The caption of figure B1 is not clear. I couldn’t understand what it means that the circles correspond to a certain number of cross sections. How is the center and radius of each circle calculated? How is the red contour calculated?
References
Lorenz, E. N. (1955). Available potential energy and the maintenance of the general circulation. Tellus, 7(2), 157-167.
Okajima, S., Nakamura, H., & Kaspi, Y. (2022). Energetics of transient eddies related to the midwinter minimum of the North Pacific storm-track activity. Journal of climate, 35(4), 1137-1156. |
